An Efficient Mice Flow Routing Algorithm for Data Centers Based on Software-Defined Networking

Amezquita-Suarez, Felipe; Estrada-Solano, Felipe; Fonseca, Nelson L. S. da; Rendón, Óscar Mauricio Caicedo

doi:10.1109/icc.2019.8761552

Cited by 9 publications

(14 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Literature [9] picks the path with the lowest delay for miceflows and [10] assigns dedicated low-latency paths for them. [11] decreases the delay by reducing the number of flow rules installed to transmit mice-flows. However, all of these solutions are based on manual design which is non-intelligent, it implies that when similar traffic patterns happen, the same paths will be selected even the routing strategies have resulted in poor network performance.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Deep Q-Learning for Routing Schemes in SDN-Based Data Center Networks

Sun

Meng

et al. 2020

IEEE Access

View full text Add to dashboard Cite

In order to adapt to the rapid development of cloud computing, big data, and other technologies, the combination of data center networks and SDN is proposed to make network management more convenient and flexible. With this advantage, routing strategies have been extensively studied by researchers. However, the strategies in the controller mainly rely on manual design, the optimal solutions are difficult to be obtained in the dynamic network environment. So the strategies based on artificial intelligence (AI) is being considered. This paper proposes a novel routing strategy based on deep Q-learning (DQL) to generate optimal routing paths autonomously for SDN-based data center networks. To satisfy the different demands of mice-flows and elephant-flows in data center networks, deep Q networks are trained for them respectively to achieve low latency and low packet loss rate for mice-flows as well as high throughput and low packet loss rate for elephant-flows. Furthermore, with the consideration of the distribution of traffic and the limitated resources of data center networks and SDN, we choose port rate and flow table utilization to describe the network state. Simulation results show that compared with Equal-Cost Multipath Routing (ECMP) and SRL+FlowFit, the proposed routing scheme can reduce both the average delay of mice-flows and average packet loss rate, while increase the average throughput of elephant-flows.

show abstract

Section: Related Workmentioning

confidence: 99%

“…Elephant-flows carry large amounts of data and last for a long time, while mice-flows do the opposite. In the light of traffic characteristics, literature [2][3][4][5][6][7][8][9][10][11] are focus on the routing approaches in data center networks based on SDN. However, all of these routing strategies need to be designed manually.…”

Section: Introductionmentioning

confidence: 99%

Deep Q-Learning for Routing Schemes in SDN-Based Data Center Networks

Sun

Meng

et al. 2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…These technologies can provide performance isolation between collocated VMs to improve application performance and prevent interference attacks. However, server virtualisation itself is insufficient to address all limitations of scalability and managing the growing traffic in DCNs [1,35].…”

Section: Software-defined Networking Data Centre Networkmentioning

confidence: 99%

“…Then, the flow sizes are compared with a static and predefined threshold at the end-switch [2,62]. In Host-Based identification, when the measurement of a flow (e.g., socket buffer and flow size) exceeds a previously set threshold value, the detector determines if the flow is a HH or not [24,35,63]. It is important to highlight that the use of a static and pre-defined threshold offers a rapid identification but low accuracy when the traffic is dynamic and grows suddenly.…”

Section: Related Workmentioning

confidence: 99%

An Approach Based on Knowledge-Defined Networking for Identifying Heavy-Hitter Flows in Data Center Networks

et al. 2019

View full text Add to dashboard Cite

Heavy-Hitters (HHs) are large-volume flows that consume considerably more network resources than other flows combined. In SDN-based DCNs (SDDCNs), HHs cause non-trivial delays for small-volume flows known as non-HHs that are delay-sensitive. Uncontrolled forwarding of HHs leads to network congestion and overall network performance degradation. A pivotal task for controlling HHs is their identification. The existing methods to identify HHs are threshold-based. However, such methods lack a smart system that efficiently identifies HH according to the network behaviour. In this paper, we introduce a novel approach to overcome this lack and investigate the feasibility of using Knowledge-Defined Networking (KDN) in HH identification. KDN by using Machine Learning (ML), allows integrating behavioural models to detect patterns, like HHs, in SDN traffic. Our KDN-based approach includes mainly three modules: HH Data Acquisition Module (HH-DAM), Data ANalyser Module (HH-DANM), and APplication Module (HH-APM). In HH-DAM, we present the flowRecorder tool for organizing packets into flows records. In HH-DANM, we perform a cluster-based analysis to determine an optimal threshold for separating HHs and non-HHs. Finally, in HH-APM, we propose the use of MiceDCER for routing non-HHs efficiently. The per-module evaluation results corroborate the usefulness and feasibility of our approach for identifying HHs.

show abstract

“…An elephant data is a single large chunk of data whereas mice flows are smaller data packets with a higher priority. When a large number of mice flows arrive at the network, the elephant flow even though smaller than the total mice data chunks have to wait, causes alteration in the flow table thereby reducing the overall performance and induced delay in the network 19,20 . Similarly, the performance corruption of mouse flows, caused by elephant flows also happens.…”

Section: Introductionmentioning

confidence: 99%

Fault tolerant resource allocation in fog environment using game theory‐based reinforcement learning

Divya

2021

Concurrency and Computation

View full text Add to dashboard Cite

Realtime decision making is associated with an on-demand, latency aware resource allocation. Fog nodes along with cloud infrastructure, when used effectively can ensure real-time decision making. In this article, we propose an efficient resource allocation and fault tolerance mechanism for the fog layer. Our work takes the advantage of game theory, where Nash equilibrium is the initial allocation strategy, which is then passed on to the reinforcement learner. The allocation is done proactively based on the network status and traffic history. The performance of our system is compared with the existing open shortest path first and neural network algorithms. Besides, fault tolerance mechanism has also been proposed which takes the advantage of the fail-over cluster formation to find the link failure and provide an alternate path in the smart switch, which is the networking component of the fog network. The proposed work gives an improved recovery time and average service time in case of failure with a recovery time of 32 ms. The experimental results are justified in terms of improved service time, lower delay, and optimal energy utilization. K E Y W O R D Sdeep reinforcement learning, failover cluster, fog computing, game theory, resource allocation INTRODUCTIONA surge in the importance of sensors and real-time monitoring has brought about the convergence of two major technologies, the cloud and Internet of Things (IoT), which lead to the emergence of cloud of Things. 1 Processing of large streaming data and providing low latent, power-efficient, high accurate decision making has yet again emerged a new technology-Fog Computing, proposed by CISCO. 2,3 This computing layer helps to alleviate congestion in the cloud by the execution of time-sensitive tasks and data filtering near the edge. 4 Researchers use the terms Edge/fog/mist computing based on the distance of the computing layer from the source of data. 5 Our work takes the advantage of the fog layer to execute the latency aware tasks and effectively distribute the load in the fog layer.One of the most commonly used techniques for load distribution is task or code offloading 6 which helps to distribute the load from the cloud to the lower layer. Latency sensitive tasks are offloaded to the intermediate layer near the edge for better performance. 7,8 This helps in a significant reduction of data being transferred to a remote location, thereby saving bandwidth utility.Software-defined network (SDN) plays a major role as the networking component in a fog network creation. The SDN is an intelligent paradigm of networking, which consists of two planes, the data and the control plane to take care of delivery and networking of the application. 9,10 The Open-Flow protocol ensures the communication between the control plane and the data plane, 11 which instructs the data plane reactively or proactively on packet handling.Traffic engineering plays a key role in path-finding and route establishment in a network. A good algorithm enhances the throughput and minimizes the d...

show abstract

An Efficient Mice Flow Routing Algorithm for Data Centers Based on Software-Defined Networking

Cited by 9 publications

References 14 publications

Deep Q-Learning for Routing Schemes in SDN-Based Data Center Networks

Deep Q-Learning for Routing Schemes in SDN-Based Data Center Networks

An Approach Based on Knowledge-Defined Networking for Identifying Heavy-Hitter Flows in Data Center Networks

Fault tolerant resource allocation in fog environment using game theory‐based reinforcement learning

Contact Info

Product

Resources

About